Thursday, 2017-11-23

*** hongbin has quit IRC00:17
*** yangyapeng has quit IRC00:23
*** yangyapeng has joined #heat00:24
*** mwynne has joined #heat00:25
*** yangyapeng has quit IRC00:28
*** zhurong has joined #heat00:46
*** masber has quit IRC00:47
*** fragatin_ has joined #heat01:02
*** fragatina has quit IRC01:06
*** fragatin_ has quit IRC01:06
*** Guest60073 has quit IRC01:16
*** Guest60073 has joined #heat01:18
kazsh ricolin: thanks! the release note job error has been cleared, https://review.openstack.org/#/c/520889/01:59
*** AlexeyAbashkin has joined #heat02:08
*** wangbo has joined #heat02:09
*** yangyapeng has joined #heat02:11
*** AlexeyAbashkin has quit IRC02:12
*** Qiming has quit IRC02:16
*** Qiming has joined #heat02:17
*** masber has joined #heat02:38
*** masuberu has joined #heat02:42
*** masber has quit IRC02:47
*** ramishra has joined #heat02:54
*** wangbo has quit IRC03:02
*** AlexeyAbashkin has joined #heat03:09
openstackgerritRabi Mishra proposed openstack/heat master: WIP Native devstack functional jobs  https://review.openstack.org/52123603:10
*** ricolin has joined #heat03:12
*** Qiming has quit IRC03:12
*** wangbo has joined #heat03:13
*** AlexeyAbashkin has quit IRC03:13
*** Qiming has joined #heat03:14
*** zhurong has quit IRC03:24
*** zhurong has joined #heat03:28
openstackgerritRabi Mishra proposed openstack/heat master: WIP Native devstack functional jobs  https://review.openstack.org/52123603:38
*** zhurong has quit IRC03:47
*** mdnadeem has joined #heat03:51
openstackgerritRabi Mishra proposed openstack/heat master: WIP Native devstack functional jobs  https://review.openstack.org/52123604:12
*** flwang1 has quit IRC04:14
*** gkadam has joined #heat04:36
*** wangbo has quit IRC04:41
*** wangbo has joined #heat05:19
*** zhurong has joined #heat05:24
*** k_mouza has joined #heat05:47
*** hjensas has quit IRC05:47
*** k_mouza has quit IRC05:52
*** skramaja has joined #heat05:59
*** moshele has joined #heat06:31
openstackgerritRabi Mishra proposed openstack/heat master: WIP Native devstack functional jobs  https://review.openstack.org/52123606:34
mosheleramishra: is the failed test is now passing https://review.openstack.org/#/c/521506/ ? I think all the failure are not related to the patch06:37
ramishramoshele: yeah looks like06:38
openstackgerritHuangsm proposed openstack/heat master: ForceDelete Instance  https://review.openstack.org/45494106:44
*** kiennt26 has joined #heat06:55
*** gkadam has quit IRC06:59
*** kei-ichi has quit IRC07:01
*** kei-ichi has joined #heat07:02
mosheleramishra: can I get you blessing to merge it in nova07:07
ramishramoshele: sure, if you mean I +1 your patch:)07:11
*** jprovazn has joined #heat07:17
*** jtomasek has joined #heat07:38
*** flwang1 has joined #heat07:40
*** wangbo has quit IRC07:41
*** wangbo has joined #heat07:43
openstackgerritRabi Mishra proposed openstack/heat master: WIP Native devstack functional jobs  https://review.openstack.org/52123607:43
*** rcernin has quit IRC07:44
openstackgerritHuangsm proposed openstack/heat master: ForceDelete Instance  https://review.openstack.org/45494107:47
*** wangbo has quit IRC07:54
*** shardy has joined #heat08:02
*** armaan has joined #heat08:05
*** AlexeyAbashkin has joined #heat08:06
*** yangyapeng has quit IRC08:06
*** yangyapeng has joined #heat08:07
*** wangbo has joined #heat08:27
*** jistr|off|trng is now known as jistr08:28
*** gkadam has joined #heat08:28
*** dbecker has joined #heat08:28
*** amoralej|off is now known as amoralej08:37
*** jprovazn has quit IRC08:47
*** maestropandy has joined #heat09:10
*** sdaniel has joined #heat09:17
sdanielHi, I would like to ask some advice or suggestion about a scale down 'problem'09:18
sdanielIf somebody could help me about it, that would be great09:18
openstackgerritKien Nguyen proposed openstack/heat master: Change transport_url in grenade-multinode job  https://review.openstack.org/51040009:22
therveramishra, I thought we kept functional tests in-tree?09:23
thervesdaniel, Sure go ahead09:24
ramishratherve: Any reason for that? As we expect users to run all tests with tempest, I think we agreed to move all09:27
therveramishra, Well I didn't agree :)09:27
sdanielI have a university project where I have to improve the following situation: Normal way behind a load balancer, if an autoscaling group reach thre treshold for scale down (like avg cpu usage less then X%), then then heat will destroy the latest generated instance(s)09:27
sdanielBut09:27
therveramishra, Functional tests make no sense on old releases. When we fix a bug and write a regression test, it's not meat to pass 2 releases ago09:27
therveAlso, it makes it harder to write bug fixes09:27
therveI'm pretty sure that's what we agreed upon at the PTG, but you weren't there09:28
ramishratherve: I think other plugins have scenario tests09:28
therveramishra, Yeah scenarios are fine. functional is the one I want to keep in-tree09:29
ramishratherve: What's the difference?09:29
sdanielIt is possible that instance has the highest network usage, because it hosts a webserver for videos. During the websession, the Load balancer wont get new request from the client, and the cleint will lose the connection and the video will stop09:29
therveramishra, scenarios are not bug/regression tests, they are feature tests09:29
therveramishra, FTR see https://etherpad.openstack.org/p/heat-queens-ptg-tempest-plugin09:30
therve"Create the new repo and import API and scenarios tests"09:30
therve"Make the functional tests in heat tree reuse the code from the plugin repository"09:30
*** efoley has joined #heat09:30
ramishratherve: I think we discussed about it later and agreed to move all, not sure if you were not there during those discussions.09:31
ramishraIf we break functional tests we would break api tests too09:31
thervebreak?09:31
therveI'm talking about new tests, not breaking existing tests09:32
*** zhurong has quit IRC09:32
*** wangbo has quit IRC09:32
ramishraI mean if we add a feature like https://review.openstack.org/#/c/509760/09:32
ramishraotherwise it should be manageable09:32
therveYeah that one was a bad example09:33
ramishraNew tests would be for new features or existing features09:33
ramishraAnd if it's for existing features it would work09:33
ramishraif it's new features we've to find a way to run on > specific release09:34
therveramishra, See https://review.openstack.org/#/c/509921/ for example what I mean09:34
therveThere is no way that test works on deployed heat09:34
*** gfidente has joined #heat09:34
*** gfidente has quit IRC09:34
*** gfidente has joined #heat09:34
ramishraWe're not changing to use the plugin for older releases like pike, that's my way of looking at it09:35
ramishraFor new releases when we add new tests it would be linked to a feature so we would run api/functional tests for releases beyond that09:36
ramishraOtherwise we can't make the plugin branchless09:37
therveMokay.09:37
ramishraOther plugins and tests in tempest repo use api microvcersions09:38
therveI still think the workflow for bug fixing is horrible then09:38
therveYeah but we don't have that09:38
*** rcernin has joined #heat09:38
ramishraI mean that's the reason I never wanted the split but we had no say09:38
ramishraI would not prefer we maintain some tests in-tree and some in the plugin09:38
therveOf course we have a say :)09:39
ramishraAnd we should not break existing tests and I mentioned that in the mail, but it would work if we don't have heat functional jobs for the tempest-plugin09:40
ramishraIf we don't have any fucntional jobs in heat-tempest-plugin, breaking tests would still work and it can be tested with depends on, where the test change merges first09:40
therveTo be fair we shouldn't do that09:42
therveAt least not too often09:42
therveThe last one was borderline, as we could have written the tests to work with both responses09:43
ramishrayes, it would stop us to do those kind of changes, which is good IMO09:45
* therve shrugs09:46
therveFrankly I'm mostly annoyed that we spent time discussing in person at the PTG, to then ignore the discussion09:49
therveThose meetings seem quite useless09:49
ramishratherve: tbh, after PTG we discussed about what tests to keep/move in few weekly meetings too.09:56
ramishraI would not prefer few tests here and few there09:57
therveramishra, Sure, but I don't need to travel 20h then :)09:57
ramishrabut anyway we can still change if we want09:57
therveWell the merge didn't work09:57
therveSo there is that :)09:58
ramishratherve: I think that's probably some othe reason may be https://review.openstack.org/#/c/519578/ has to go in, I don't know09:59
*** dbecker has quit IRC09:59
therveMaybe yeah10:01
ramishraIMO, this is about trying out and getting to know what works best and decisions can be changed if required, whether taken in PTG or now10:02
therveI'm not arguing that decisions shouldn't be changed.10:07
therveI'm arguing that we don't need to meet in person.10:07
sdanielHi, sorry for interrupt you. Any suggestion about my question ?10:12
*** armaan has quit IRC10:16
*** armaan has joined #heat10:17
thervesdaniel, Did you talk about that one a few days ago?10:20
sdanielYes, but I think we did not finish the discussion10:24
*** k_mouza has joined #heat10:24
therveOh I finished it :)10:26
thervesdaniel, You should look into something else, heat won't answer your problem10:26
thervesdaniel, Maybe senlin, I think it has more flexible options wrt to deletion10:27
*** jprovazn has joined #heat10:29
*** yangyapeng has quit IRC10:31
*** armaan has quit IRC10:31
*** armaan has joined #heat10:31
*** yangyapeng has joined #heat10:31
sdanielBecause of it a university project, if it really can't solve in heat, I have to explain why.10:31
therveThere are various answers to that. Maybe the simpler is that Heat autoscaling support isn't good.10:33
sdanielOtherwise I have to give opportunites, which idea could work in heat, and aftr that explain why it is not a proper and acceptable way10:33
sdanielI know that, it does not support it at now.10:34
therveWith the current autoscaling support, you can't treat a node differently10:34
therveYou can't chose what will be deleted in a scale down10:34
therveSo your scenario is just not supported10:35
*** wangbo has joined #heat10:35
sdanielI know, I know10:35
sdanielBut if there will be a possibilites, to improve the heat functionalities from my side10:36
sdanielJust show to my university an example that it can be solved10:36
sdanielThat would be an achieved goal from my side.10:37
therveWell you need a way to express "don't delete this node"10:38
therveAWS has a way to do that10:39
therveI don't know what it would mean to implement it in heat though10:39
*** yangyapeng has quit IRC10:40
thervesdaniel, Otherwise, senlin uses heat, so it's a good thing to look at10:40
sdanielOk, I will check senlin too10:43
*** kiennt26 has quit IRC10:43
sdanieltherve, what If I'm trying to detect, which node should be deleted ?10:44
sdanielSomehow heat can use this information during the "resize" ?10:45
therveThere is no interface to do this right now10:46
*** moshele has left #heat10:51
*** maestropandy has quit IRC10:51
*** Guest60073 has quit IRC10:53
flwang1therve: sorry for chiming in, i have a question when using magnum + heat10:58
flwang1therve: i'm using magnum Pike and heat Kilo, and I got this error http://paste.openstack.org/show/627213/10:58
*** yangyapeng has joined #heat11:05
*** wangbo has quit IRC11:05
*** armaan has quit IRC11:06
*** armaan has joined #heat11:07
*** yangyapeng has quit IRC11:13
*** armaan has quit IRC11:14
*** armaan has joined #heat11:15
*** jprovazn has quit IRC11:18
*** jprovazn has joined #heat11:19
openstackgerritMerged openstack/python-heatclient master: Don't preformat stack output show list/map values  https://review.openstack.org/52196911:21
*** armaan has quit IRC11:24
*** armaan has joined #heat11:25
*** yangyapeng has joined #heat11:27
*** wangbo has joined #heat11:28
*** yangyapeng has quit IRC11:31
flwang1zbitter:  i'm using magnum Pike and heat Kilo, and I got this error http://paste.openstack.org/show/627213/  any idea?11:32
*** yangyapeng has joined #heat11:33
ramishraflwang1: I think the server_image parameter is incorrect https://github.com/openstack/magnum/blob/stable/pike/magnum/drivers/k8s_fedora_atomic_v1/templates/kubemaster.yaml#L50311:40
ramishraflwang1: btw, why are you using kilo heat, it's not supported since long time11:40
ramishraflwang1: I think it would be better for you to check with the magnum team11:41
flwang1ramishra: just because it's sitting in our cloud, and we're trying magnum, i know we should upgrade it anyway, but I'd like to continue my test before the upgrade, that's the background11:41
strigaziramishra: what is incorrent in the above file?11:41
flwang1ramishra: strigazi is the PTL of magnum11:42
ramishrastrigazi: which file?11:42
strigaziramishra: https://github.com/openstack/magnum/blob/stable/pike/magnum/drivers/k8s_fedora_atomic_v1/templates/kubemaster.yaml#L50311:42
ramishrastrigazi: I just metioned where it's validating that property picked from the parameter11:43
ramishraand it seems to be using 'id' instead of a valid image id/name11:43
strigaziramishra: to test quickly we tried with this template: http://paste.openstack.org/raw/627207/11:44
strigazishouldn't this ^^  work11:45
ramishraI don't see server_image parameter passed to 'openstack stack create'11:46
ramishraoh sorry it's there11:46
ramishraat the end11:47
ramishrait should work11:47
flwang1ramishra: so can i get some ideas where should i add log/breakpoint to get more info?11:48
strigaziramishra: can we use glance v2 with heat kilo? is there such an option?11:51
flwang1strigazi: i  think it depends if that's the root cause11:52
ramishraI think you can check here https://github.com/openstack/heat/blob/kilo-eol/heat/engine/clients/os/glance.py#L6111:52
ramishrawhat you get there from glance11:52
ramishrastrigazi: I think it's using v1 https://github.com/openstack/heat/blob/kilo-eol/heat/engine/clients/os/glance.py#L5011:53
flwang1ramishra: ok11:53
flwang1ramishra: yep, it's hardcode with v1, but it should work11:53
*** dbecker has joined #heat11:54
ramishrakilo is too old and not supported;) I would not use pike magnum with kilo heat.11:55
*** gkadam_ has joined #heat11:55
*** gkadam has quit IRC11:57
openstackgerritHuangsm proposed openstack/heat master: ForceDelete Instance  https://review.openstack.org/45494111:58
flwang1ramishra: thanks, i totally agree11:59
*** efoley has quit IRC11:59
flwang1but i just don't want to stop my testing because we're also mixing the other services version, i need to figure out if all of them can work together11:59
flwang1that's the hard part :(11:59
therveYou figured it out? It can't12:01
*** amoralej is now known as amoralej|lunch12:07
*** k_mouza has quit IRC12:08
flwang1therve: what do you mean 'it can't'?12:09
therveflwang1, Well using pike magnum doesn't work with kilo heat?12:10
flwang1therve: i think it doesn't work ;)12:11
therveYeah that's what I just said12:11
flwang1but i just want to try to workaround it to continue my test12:11
therveThat's not a good idea though :)12:11
flwang1haha, true12:11
flwang1i have just workaround above issue, by simply return the image id instead of talk with glance again to get it12:12
*** k_mouza has joined #heat12:12
flwang1ramishra: therve: strigazi: thank you for all your help12:13
therveAlso please really make it a priority to upgrade your heat deployed version :)12:14
*** k_mouza has quit IRC12:17
*** rcernin has quit IRC12:17
openstackgerritThomas Herve proposed openstack/heat-agents master: Make chef agent tests more reliable  https://review.openstack.org/52253412:26
flwang1therve: hah, will do12:32
*** k_mouza has joined #heat12:35
*** k_mouza has joined #heat12:35
*** efoley has joined #heat12:36
*** yassine has joined #heat12:39
*** yassine is now known as Guest5790612:39
openstackgerritMerged openstack/heat-agents master: Retry logic for url request in heat-config-notify  https://review.openstack.org/51941712:41
*** k_mouza has quit IRC12:41
*** k_mouza has joined #heat12:42
*** k_mouza has quit IRC12:45
*** k_mouza has joined #heat12:46
*** derekh has joined #heat12:57
*** ramishra has quit IRC12:58
*** wangbo has quit IRC13:00
*** jmlowe has quit IRC13:03
*** Hazelesque has quit IRC13:07
*** yangyapeng has quit IRC13:08
*** yangyapeng has joined #heat13:09
*** yangyapeng has quit IRC13:13
*** k_mouza has quit IRC13:14
*** efoley has quit IRC13:14
*** k_mouza has joined #heat13:14
*** efoley has joined #heat13:19
*** Hazelesque has joined #heat13:19
*** blues-man has joined #heat13:29
*** dprince has joined #heat13:31
*** k_mouza has quit IRC13:38
*** ramishra has joined #heat13:40
*** skramaja has quit IRC13:44
*** yangyapeng has joined #heat13:46
*** k_mouza has joined #heat13:46
*** yangyape_ has joined #heat13:47
openstackgerritMerged openstack/heat master: Ignore errors in OS::stack_id output  https://review.openstack.org/50724913:47
*** k_mouza has quit IRC13:49
*** k_mouza has joined #heat13:49
*** yangyapeng has quit IRC13:51
*** amoralej|lunch is now known as amoralej13:52
*** efoley has quit IRC13:55
*** efoley has joined #heat13:56
*** dprince has quit IRC14:00
*** ramishra has quit IRC14:00
*** ramishra has joined #heat14:02
*** armaan_ has joined #heat14:19
*** armaan has quit IRC14:19
*** blues-man has quit IRC14:20
*** k_mouza has quit IRC14:25
*** k_mouza has joined #heat14:32
*** armaan_ has quit IRC14:32
*** dbecker has quit IRC14:35
*** blues-man has joined #heat14:39
*** mdnadeem has quit IRC14:48
*** dbecker has joined #heat14:48
*** pbourke has quit IRC14:50
*** pbourke has joined #heat14:52
*** sdaniel has quit IRC14:54
*** k_mouza has quit IRC14:57
*** k_mouza has joined #heat15:06
*** yangyape_ has quit IRC15:27
*** bluesman_ has joined #heat15:27
*** yangyapeng has joined #heat15:29
*** blues-man has quit IRC15:30
*** szaher has quit IRC15:34
*** szaher has joined #heat15:35
*** bluesman_ has quit IRC15:35
*** gkadam_ has quit IRC15:38
*** gkadam has joined #heat15:40
*** ramishra has quit IRC15:45
*** bluesman_ has joined #heat15:50
*** gkadam is now known as gkadam-dinner15:53
*** serlex has joined #heat15:56
*** serlex has quit IRC16:10
*** k_mouza has quit IRC16:25
*** k_mouza has joined #heat16:30
*** huangtianhua_ has quit IRC16:49
*** huangtianhua_ has joined #heat16:50
*** AlexeyAbashkin has quit IRC16:51
*** bluesman_ has quit IRC16:58
*** k_mouza has quit IRC17:17
*** gkadam-dinner is now known as gkadam17:22
*** gkadam has quit IRC17:27
*** Guest57906 has quit IRC17:35
*** jprovazn has quit IRC17:41
*** dbecker has quit IRC17:45
*** tellesnobrega has joined #heat17:53
*** derekh has quit IRC17:55
openstackgerritRico Lin proposed openstack/heat master: Translate networks and subnet in cluster template  https://review.openstack.org/50733117:59
tellesnobregaricolin, hey, can I ask you a couple question on heat templates?18:00
ricolintellesnobrega, sure18:07
tellesnobregaI'm working with cluster scale down now, and what it basically does it update the heat template with the number of instances on the node group and the stack will update the cluster by removing the last created instance from the node group. What I'm trying to do is to select the instances that will be deleted.18:08
*** shardy has quit IRC18:09
tellesnobregado you know of any examples of similar stuff? so I can take a look on how the stack can be done18:10
ricolinAdd that resource to your black list and it will remove that instance when scale down18:11
ricolinin resource group properties add       removal_policies:18:12
ricolin        - resource_list:18:12
ricolin          - "2"18:12
ricolin18:12
ricolin      removal_policies:18:12
ricolin        - resource_list:18:12
ricolin          - "2"18:12
ricolin2 is the index in resource group that I want to delete18:13
ricolintellesnobrega, ^^^18:13
tellesnobregaricolin, thanks18:13
*** ricolin has quit IRC18:23
*** efoley has quit IRC18:29
*** amoralej is now known as amoralej|off18:39
*** tellesnobrega has quit IRC18:55
*** tellesnobrega has joined #heat18:57
*** jmlowe has joined #heat19:02
*** Guest57906 has joined #heat19:05
*** flwang1 has quit IRC19:48
*** k_mouza has joined #heat20:46
*** k_mouza has quit IRC20:50
*** gfidente has quit IRC20:53
*** fragatina has joined #heat20:55
*** fragatina has quit IRC20:56
*** fragatina has joined #heat20:56
*** hongbin has joined #heat21:16
*** jmlowe has quit IRC21:25
*** flwang1 has joined #heat21:31
*** efoley has joined #heat21:36
*** efoley has quit IRC21:36
*** efoley has joined #heat21:36
*** liusheng has quit IRC21:47
*** liusheng has joined #heat21:48
*** ccamacho has quit IRC21:55
*** k_mouza has joined #heat22:01
*** rcernin has joined #heat22:37
*** efoley has quit IRC22:41
-openstackstatus- NOTICE: Zuul has been restarted due to an unexpected issue. We're able to re-enqueue changes from check and gate pipelines, please check http://zuulv3.openstack.org/ for more information.22:53
*** fragatin_ has joined #heat22:56
*** fragatina has quit IRC22:58
*** k_mouza has quit IRC23:04
*** pbourke has quit IRC23:13
*** pbourke has joined #heat23:19
ivveanyone here using python-flameclient ?23:42

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!